lvill INTRODUCTION. 



All pressures are expressed in heights of mercurial column under stan- 

 dard gravity. 



The important advance made since the time of Regnault consists in 

 recognizing that the value of A differs materially according to whether the 

 wet-bulb is in quiet or moving air. This was experimentally demonstrated 

 by the distinguished Italian physicist, Belli, in 1830, and was well known 

 to Espy, who always used a whirled psychrometer. The latter describes 

 his practice as follows: "When experimenting to ascertain the dew-point 

 by means of the wet-bulb, I always swung both thermometers moderately 

 in the air, having first ascertained that a moderate movement produced 

 the same depression as a rapid one." 



The principles and methods of these two pioneers in accurate psychrom- 

 etry have now come to be adopted in the standard practice of meteorolo- 

 gists, and psychrometric tables are adapted to the use of a whirled or ven- 

 tilated instrument. 



The factor A depends in theory upon the size and shape of the ther- 

 mometer bulb, largeness of stem and velocity of ventilation, and different 

 formulae and tables would accordingly be required for different instruments. 

 But by using a ventilating velocity of three meters or more per second, the 

 differences in the results given by different instruments vanish, and the 

 same tables can be adapted to any kind of a thermometer and to all changes 

 of velocity above that which gives sensibly the greatest depression of the 

 wet-bulb temperature; and with this arrangement there is no necessity to 

 measure or estimate the velocity in each case further than to be certain that 

 it does not fall below the assigned limit. 



The formula and tables here given for obtaining the vapor pressure 

 and dew-point from observations of the whirled or ventilated psychrom- 

 eter are those deduced by Prof. Wm. Ferrel (Annual Report Chief 

 Signal Officer, 1886, Appendix 24) from a discussion of a large number of 

 observations. 



Taking the psychrometric formula in metric units, pressures being 

 expressed in millimeters and temperatures in centigrade degrees, Prof. Ferrel 

 derived for A the value 



A = 0.000656 (1 +0.0019 t'). 



In this expression for A, the factor depending on t' arises from a similar 

 term in the expression for the latent heat of water, and the theoretical value 

 of the coefficient of t' is 0.00115. Since it would require a very small change 

 in the method of observing to cause the difference between the theoretical 

 value and that obtained from the experiments, Prof. Ferrel adopted the 

 theoretical coefficient 0.00115 and then recomputed the observations, ob- 

 taining therefrom the final value 



A = 0.000660 (1 +0.00115 /')• 



